GitHub MCP Server for Cursor IDE

What is GitHub MCP Server for Cursor IDE?

It is a GitHub integration server that allows you to manage GitHub repositories directly from Cursor IDE. It provides a RESTful API to interact with GitHub's API for creating and managing repositories, intended for developers using Cursor IDE.

How to use GitHub MCP Server for Cursor IDE?

Clone the repository, install dependencies with npm install, create a .env file with a GitHub Personal Access Token and other configurations, then start the server with npm run dev. The server runs on http://localhost:3000 by default. Use endpoints like POST /api/github/repos to create repositories and GET /api/github/repos to list them.

Key features of GitHub MCP Server for Cursor IDE

• Create new GitHub repositories via RESTful API
• List existing repositories accessible to the user
• Secure token‑based authentication with rate limiting
• Comprehensive error handling and logging
• Health check endpoint at /health

Use cases of GitHub MCP Server for Cursor IDE

• Quickly create a new GitHub repository from Cursor IDE
• Browse your personal or organization repositories without leaving the editor
• Automate repository creation as part of a development workflow
• Integrate GitHub repository management into Cursor IDE’s MCP ecosystem

FAQ from GitHub MCP Server for Cursor IDE

What dependencies are required?

Node.js v14 or higher and npm v6 or higher. A GitHub account with a Personal Access Token is needed.

How do I generate a GitHub Personal Access Token?

Go to GitHub.com → Settings → Developer Settings → Personal Access Tokens → Tokens (classic), generate a new token with repo, read:org, and admin:repo_hook scopes, then paste it into the .env file.

What endpoints does the server expose?

GET /health (health check), GET /api/github/repos (list repos), and POST /api/github/repos (create repo). The POST endpoint accepts a JSON body with name, description, private, and auto_init fields.

What security measures are implemented?

Token‑based authentication, rate limiting, input validation, secure environment variable handling, CORS protection, and Helmet security headers.

Where are logs stored?

Logs are stored in error.log (error‑level) and combined.log (all logs).